Active Parameters
-
Context Length
1M
Modality
Multimodal
Architecture
Mixture of Experts (MoE)
License
-
Release Date
30 Jun 2026
Knowledge Cutoff
Jan 2026
Rank
#22
| Benchmark | Score | Rank |
|---|---|---|
Agentic Coding high | 0.59 | 11 |
Agentic Coding max | 0.59 | 11 |
Agentic Coding Standard | 0.59 | 11 |
Agent Arena high | 0.08 | 11 |
Reasoning high | 0.89 | 14 |
Reasoning max | 0.89 | 14 |
Reasoning Standard | 0.89 | 14 |
Coding high | 0.81 | 15 |
Coding max | 0.81 | 15 |
Coding Standard | 0.81 | 15 |
Agentic Index max | 0.50 | 15 |
Mathematics high | 0.93 | 18 |
Mathematics max | 0.93 | 18 |
Mathematics Standard | 0.93 | 18 |
Intelligence Index max | 0.55 | ⭐ 18 |
Coding Index max | 0.71 | 18 |
General high | 0.76 | 25 |
General max | 0.76 | 25 |
LiveBench Average high | 0.76 | 29 |
LiveBench Average max | 0.76 | 29 |
LiveBench Average Standard | 0.76 | 29 |
Web Development high | 1537 | 29 |
Data Analysis high | 0.72 | 39 |
Data Analysis max | 0.72 | 39 |
General Text high | 1462 | 50 |
Overall Rank
#22
Coding Rank
#20
Anthropic's Claude Sonnet 5 is built for agentic workflows, long-horizon coding, and high-efficiency reasoning with a 1M token context window and adaptive thinking by default.
Detailed architecture specifications are limited for proprietary models.
Anthropic's Claude Sonnet 5, released June 30, 2026, is built to be the most agentic Sonnet-class model, offering near-Opus 4.8 level intelligence, coding, and tool use at Sonnet pricing. It features a 1M token context window by default, 128k maximum output tokens, and adaptive thinking enabled by default.
APX AI
Online